JP2978891B1 - Head positioning control method for magnetic disk drive - Google Patents

Abstract

[PROBLEMS] Conventionally, since an asynchronous component of a runout is also included, the correction accuracy is low and the effect of suppressing a synchronous runout is low. SOLUTION: A runout processing unit 2 calculates a runout component synchronized with the rotation of the disk for each servo sector by an operation using a Fourier transform equation. The run-out processing unit 2 calculates a run-out reference correction coefficient and stores it in the memory 16. Using this memory reference correction coefficient,
The position error signal U2 can be generated instantaneously. The runout detector 17 reads a signal in the data area while performing data read / write operation by the head 9, and always detects a temporal change of the runout. Further, the detected value and the reference value are compared by a comparator, and when the detected value exceeds a preset reference value, a command to stop the read / write operation by the head 9 is output to the upper interface circuit 14. I do.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a head positioning control method in a magnetic disk drive, and more particularly to a head positioning control method in a magnetic disk drive for positioning a magnetic head at a desired track position on a magnetic disk medium.

[0002]

2. Description of the Related Art Among magnetic disk drives, a hard disk is mainly used as a storage device built in a computer due to its relatively fast data reading and writing and a large capacity. Accordingly, miniaturization and high-density magnetic disk devices are also desired. In addition to hard disks, magnetic disk devices such as removable disks and High-FDDs are also in the spotlight.

[0003] There are two methods for realizing high density, a method of increasing the linear recording density (BPI) and a method of increasing the track density (TPI). In a hard disk, a magnetic head position signal (hereinafter, also referred to as a servo signal) necessary for positioning a magnetic head (hereinafter, also simply referred to as a head) at a desired track position is an amplitude difference between a burst signal of an even track and an odd track. It is generally known to be generated by
No. 8).

[0004] The servo signal is information used as a reference when positioning the head to a desired track, and it is important that the servo signal is accurately recorded on the magnetic disk medium. To increase the TPI in recent magnetic disk drives,
There is a need for higher adjacent track pitch accuracy, that is, improved servo signal accuracy. The recording pattern of the servo signal is also called a servo pattern, and is recorded by a servo track writer (STW) before shipping the product.

Hitherto, various positioning control methods have been proposed for improving the accuracy of servo signal and head positioning (Japanese Patent Application Laid-Open Nos. 1-279474 and 4-38).
778, JP-A-6-176514, JP-A-7-249276, etc.). For example, Japanese Patent Laid-Open No. 1-2
In the conventional method described in Japanese Patent No. 79474 (the title of the invention, "head positioning method"), a head and a position shift detecting circuit take out an off-track amount from servo data of a certain track, and read the track from the off-track in one round of the track. The first and second order components of the eccentric rotation period are obtained, an off-track amount of the track is created from the data, and the head is moved based on the off-track to position the head.

In the current magnetic disk drive, a sector servo system in which servo signals are discretely arranged between a plurality of divided data areas is mainly used for the purpose of improving data format efficiency. In this sector servo method, each sector on the magnetic disk has a servo area and a data area, cylinder address information and off-track information are recorded in the servo area, and sector address information and data are recorded in the data area. You. The cylinder address information is mainly used during a seek operation to perform head positioning control, and the off-track information is mainly used during on-track to perform positioning control.

For such a sector servo system,
In addition to the fact that servo signals can only be obtained at certain fixed intervals, a reduction in the number of servo sectors, that is, a reduction in sample rate, is desired in order to increase data format efficiency. However, in the magnetic disk drive of the sector servo system, it becomes difficult to set a high control band in the head positioning control system as the sampling rate is reduced, and the compression characteristics in a low frequency region are deteriorated. Due to the decrease in the compression ratio, the head follow-up characteristic is deteriorated, the S / N of the data reproduction signal is further reduced, and high positioning accuracy cannot be secured.

Further, in the magnetic disk drive, a phenomenon called runout occurs due to radial vibration due to eccentricity of the center axis of the magnetic disk or fluctuation due to the location of a reproduced signal, and a sufficient amount of servo signal is generated. Flat characteristics are not always obtained. In particular, since a runout synchronized with the rotation of a magnetic disk having a large component (hereinafter also referred to as a synchronous runout) is generated in a low frequency region, it is very necessary to secure a compression ratio and reduce the synchronous runout in order to satisfy the device specifications. Is important.

Conventionally, in a prewrite system in which a servo signal is written on a magnetic disk on the STW and then the magnetic disk is incorporated into a spindle motor, there is an unavoidable eccentricity of several tens of μm which occurs when the magnetic disk is incorporated, and runout is reduced. Is important.

In view of the above, various positioning control systems have been proposed in the past to improve the accuracy of head positioning by reducing the synchronous runout (Japanese Patent Laid-Open No. 4-32417).
No. 3, Japanese Patent No. 2,526,305). JP-A-4-
In the conventional positioning control method of a magnetic disk device described in Japanese Patent No. 324173, in a magnetic disk device of a sector servo system, the off-track amount due to the synchronous run-out of all sectors is stored in advance, and the off-track amount of the current sector and The off-track correction data is created by superimposing the off-track amount of the sector by the synchronous run-out and off-track feed-forward control is performed by the off-track correction data.

In the conventional positioning control method of a magnetic disk drive described in Japanese Patent No. 2526305, track deviation information is sampled from a servo signal of an arbitrary track and stored in storage means before performing track following control. After the Fourier transform of the track shake information is performed, the track shake information of a predetermined frequency component is subjected to inverse Fourier transform, and the inverse Fourier transformed track shake information is supplied to the head as positioning information in a feedforward manner.
According to this conventional method, the first-order component and the second-order component of the track shake are advanced in phase and subjected to inverse Fourier transform, so that future track shake information is obtained in time, thereby predicting the track shake. It is intended to perform good track following.

[0012]

However, in the above-described positioning control method for a magnetic disk drive described in Japanese Patent Application Laid-Open No. 4-324173, it is described that the off-track amount of the synchronous run-out can be corrected. Since components are also included, there is a problem that the correction accuracy is low and the effect of suppressing the synchronous run-out is low. Patent No. 252
In the conventional positioning control method of the magnetic disk drive described in No. 6305, the runout is not always detected at the time of read / write, so that the influence of the runout at the time of read / write cannot be eliminated, and the read / write malfunctions. It cannot be prevented.

The present invention has been made in view of the above points,
It is an object of the present invention to provide a head positioning control method in a magnetic disk drive capable of correcting undulation of a servo signal caused by a runout.

It is another object of the present invention to provide a head positioning control method in a magnetic disk drive capable of automatically correcting a run-out even during read / write without requiring special hardware.

[0015]

In order to achieve the above object, a first aspect of the present invention is to receive a head position signal reproduced by a magnetic head from a magnetic disk, and to calculate the rotation of the magnetic disk by using a Fourier transform equation. A runout correction data generating unit that measures a synchronized runout component and generates runout correction data, a storage unit that stores the runout correction data, a target value input from a host device, and a head position signal reproduced from a magnetic disk. Adding means for adding the runout correction data from the storage means to the first position error signal; an actuator mechanism for controlling the position of the magnetic head to a target position using the signal obtained from the adding means as a second position error signal ; Read data from magnetic disk
Runout components are always detected during
If the run-out component is larger than the preset reference value
If determined, read / write operation by magnetic head
And a run-out detector for stopping the run-out
Retry operation after read / write operation is stopped by
When the detected runout component is higher than the preset reference value
The run-out correction data
Of runout components and runout complementation
The generation of the positive data is performed again, and the storage run
The correction data is updated.

In the present invention, the magnetic disk vibrates in the radial direction due to the eccentricity of the center axis of the magnetic disk or the like, and the swell (run-out component) of the head position signal caused by a phenomenon called run-out generated by the magnetic field is expressed by a Fourier transform equation. Automatic measurement is performed by the used calculation to create run-out correction data, and the run-out correction data is stored in a storage means, and is added to the first position error signal to be used for the second run-out correction data.
Since the positioning control of the magnetic head is performed as the position error signal, an appropriate second head position error signal corresponding to the eccentricity of the magnetic disk can be instantaneously generated.

[0017]

Further , in the present invention, since the change of the run-out amount is always detected by the run-out detection section, unexpected disturbance or a sudden increase of the run-out amount can be instantaneously detected , and unexpected disturbance can be detected. Contamination or already set
To compensate for the decrease in the accuracy of the run-out correction value.
An appropriate run-out correction value can be updated.

[0019]

[0020]

Further, in order to achieve the above object, a second
The invention is characterized in that the run-out correction data generating means divides the magnetic disk into a plurality of zones and generates the same run-out correction data in each zone. According to the present invention, the efficiency of the storage means for storing the run-out correction data can be improved, which is desirable.

Furthermore, in order to achieve the above object, a third
As in the invention, the run-out correction data generating means periodically measures the run-out component synchronized with the rotation of the magnetic disk and generates the run-out correction data, so that the run-out correction data can easily cope with aging and environmental changes. So desirable.

In order to achieve the above object, the fourth
According to the invention, a run-out detection unit includes a read amplifier for amplifying a read signal of a magnetic head, a peak hold circuit for holding a peak value of an output read signal of the read amplifier, and removing a high frequency component of an output signal of the peak hold circuit. The output signal of the low-pass filter is compared with the output reference value of the memory, and the output signal of the low-pass filter is larger than the output reference value of the memory. In one case, the comparator is configured to output an instruction signal for stopping the read / write operation by the magnetic head.

[0024]

Next, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of a head positioning control method in a magnetic disk drive according to the present invention. In this embodiment, a VCM driving current S3 is supplied to a voice coil motor (VCM) 8 which drives an actuator mechanism 10 for positioning a head 9.
, A higher-level interface circuit 14, and a lineout detector 17.

The positioning control unit 1 includes a microcomputer (hereinafter abbreviated as a microcomputer) 15, and FIG. 1 shows the microcomputer 15 in a functional block diagram. The microcomputer 15 includes a run-out processing unit 2 and addition units 3 and 4
And a compensating unit 5. In addition, the positioning control unit 1
As shown in FIG. 1, a read / write amplifier 11 for amplifying the output signal S1 of the head 9, a peak hold circuit 12 for holding the peak value of the output signal of the read / write amplifier 11, and a peak hold circuit 12 A / A which digitizes the held value and outputs it to the microcomputer 15
It includes a D converter 13, a D / A converter 6 for converting an output digital value of the compensator 5 into an analog signal, and an amplifier 7 for amplifying an output signal of the D / A converter 6 and outputting the amplified signal to the VCA 8. I have.

A discrete servo signal S2 unique to the sector servo system is obtained through the A / D converter 13, and the run-out processing unit 2 calculates a rotation synchronization error correction value U3 in consideration of the servo signal S2. Output to the interface circuit 14. The upper interface circuit 14 stores the rotation synchronization error correction value U3 in the memory 16. The compensator 5 receives the position error signal U2 from the adder 4 and
An appropriate control amount necessary for driving the M8 is output to the amplifier 7 through the D / A converter 6. The amplifier 7 amplifies the input signal and outputs to the VCM 8 a VCM drive current value S3 for moving the head 9 to a desired track position input to the upper interface circuit 14.

Further, as shown in the block diagram of FIG. 2, for example, the run-out detector 17 of FIG. 1 includes a read amplifier 18 for detecting a change in the run-out amount based on the output signal S1 of the head 9 performing the positioning operation. , A peak hold circuit 19, a low-pass filter 20 for removing noise, a memory 21 for storing a reference value of the runout amount in advance, and a comparator for comparing this reference value with the runout detection signal. (Comparator) 22.

Next, the operation of the embodiment of the present invention will be described. In the present embodiment, it is assumed that servo signals are obtained discretely (arbitrary number N) at certain fixed period intervals in consideration of the efficiency of the memory 16 for storing the runout correction value.

First, the creation of the run-out correction data will be described. The run-out processing unit 2 in the positioning control unit 1 in FIG. 1 calculates a run-out component (undulation component) synchronized with the rotation of the disk for each servo sector by an operation using a Fourier transform equation. Here, FIG.
The method of extracting the primary run-out component generated at the rotation frequency of the disk 23 will be described in detail below.

Using a general Fourier series, a first-order component and a DC (direct current) component of the runout are obtained. The coefficient value of the expansion of the Fourier series is an approximation that minimizes the mean square error. Has been proven. From the viewpoint of periodicity, signals can be classified into periodic signals and non-periodic signals. A periodic signal repeats the same waveform at a certain fixed period. If the period of the signal X (t) is P, X (t) = X (t + nP) (1) Thus, the signal value of one period from the arbitrary time t ₀ to describe a periodic signal, X (t), it is sufficient if you give the _{t 0 ≦ t <t 0 +} P (2), and the periodic signal is Fourier Can be expanded to a series.

Here, the runout is also a periodic signal of a certain period P except for the asynchronous component. If the primary component of the runout is represented by x (t), x (t) is a constant term corresponding to the DC component. And the sum of the sinusoidal signals of the period P can be approximated by the following equation.

X (t) ≒ a ₀ + a ₁ cos ωt + b ₁ sin ωt (3) where the coefficients a ₀ , a ₁ and b ₁ in the equation (3) are represented by the following equations, respectively.

[0033]

(Equation 1) Here, ω is an angular frequency of the primary run-out component, that is, what is called a disk rotation basic angular frequency.
2π / P. Further, N is a number of detected servo signal S1 obtained per round of the track, .omega.t _k is a reference index becomes 2π in one revolution. Also, cos and s
The value of "in" is stored in a table, and the values of these trigonometric functions are counted from the index to identify the number and the table is searched.

Further, second, third,. . . , Approximation accuracy can be improved by including up to the nth harmonic component,
Theoretically, when n → ∞, the signal x (t) can be completely matched. In this case, the run-out component is represented by the following equation.

[0035]

(Equation 2) Next, a procedure for creating a correction value in the run-out processing unit 2 will be described with reference to the flowchart in FIG. The run-out processing unit 2 first measures the actual head position error amount U1 for one rotation with reference to the index on the disk 23 at the time of factory shipment such as the final inspection process (step 101), and formulas (3) to ( 6), the initial values of the correction coefficients a ₀ , a ₁ , and b ₁ are calculated, and the run-out processing unit 2
(Step 10)
2).

Next, immediately after the completion of the above measurement, an actual head position error signal U1 used at the time of normal head positioning (obtained by adding the head position information signal S2 to the target input r (t) by the adder 3). Then, the correction value x (t) obtained by the equation (3) is added through the adder 4 to generate a head position error signal U2 (step 103), and normal head positioning control is performed (step 104). Next, considering the convergence of the transient response due to the addition of the correction value x (t), the process waits for 1/3 rotation (step 105). However, this waiting time is set arbitrarily.

Next, the second measurement is started at the time of actual reading / writing (step 106). However, since the start position may be different from the first measurement due to the consideration of the waiting time, it is necessary to pay attention to the table drawing.
By the second measurement, each correction coefficient stored in the memory in the run-out processing unit 2 is updated and stored again (step 107).

Similarly, the third measurement and updating of each run-out correction coefficient are performed (step 108). Thereafter, it is determined whether or not the residual position error amount, that is, the actual head position error amount U1 is equal to or smaller than an arbitrary set value (step 109). If it is larger than the arbitrary set value, measurement and updating of each run-out correction coefficient are further performed. After that (step 110), it is determined again whether the residual position error amount is equal to or less than an arbitrary set value (step 1).
09), if the residual position error is equal to or less than the arbitrary set value, the measurement is terminated (step 111).

Hereinafter, an example of the algorithm of the above calculation procedure will be described. First, the first measurement will be described.

The correction coefficient initial _{value: a 0 = a 1 = a} 2 = 0 (8) off-track amount at t _k a (remaining seek length) Rtrk
If (t _k ), Rtrk (t _k ) = x (t _k ), and a value indicating the number of the sector from the index is sect
_nm and the number N (for example, 60) of servo signals obtained per track circumference, the above-described correction coefficients a ₁ and b ₁ are expressed by the following equations, respectively.

A ₁ = a ₁ + Rtrk (t _k ) · cos (2π · sect_nm / N) (9) b ₁ = b ₁ + Rtrk (t _k ) · sin (2π · sect_nm / N) (10) One cycle of the formula based on the index,
That is, N lower-orders are repeated. After repeating for one cycle, the values of Expressions (4) to (6) are obtained. Here, it should be noted that the obtained a ₁ and b ₁ are rewritten as a ₁ data and b ₁ data.

A ₁ data = b ₁ data = 0 (11) a ₁ data = a ₁ data + (2a ₁ / N) (12) b ₁ data = b ₁ data + (2b ₁ / N) (13) The coefficients are obtained by a series of calculations.

Subsequently, the first measurement is corrected using the equation (3). The correction value x is added to the actual position error signal U1.
(T) is added, and this is regarded as a position error signal U2 at the time of control. It uses a relational expression of the following actual remaining seek length P _a (t _k).

P _a (t _k ) = P _a (t _k ) + a ₁ data · cos (2π · sect_nm / N) + b ₁ data · sin (2π · sect_nm / N) (14) Next, a similar procedure is performed. The second and subsequent measurements and corrections are performed. However, it should be noted that the new coefficient value obtained in the second measurement is updated after being added to the coefficient value obtained in the first measurement. As described above, the run-out processing unit 2 calculates the run-out reference correction coefficient and stores it in the memory 16 which is a predetermined location in FIG.

Next, data stored in the memory 16 will be described. As described above, in the magnetic disk drive of the present embodiment, the obtained servo signals are discrete, and the number is N per track. Therefore, when calculating the run-out correction value in the run-out processing unit 2, it is sufficient to generate a cos table and a sin table in increments of 360 / N degrees. At the time of the actual positioning, as can be seen from the equation (14), the correction value is added at every certain discrete cycle.

As described above, the run-out correction data to be recorded at a certain discrete period is generated, which is N different data on the same track. Further, as shown in FIG. 4, in order to increase the efficiency of the memory,
The disk 23 is divided into arbitrary M zones 25 _{1 to} 25 _M , and the same run-out correction value is used in the same zone. Further, these series of measurements and calculations are executed in the final inspection step and the like, and the obtained correction data is stored in the memory 16.

Next, the actual operation will be described. According to equation (14), the runout correction value U3 is added to the actual position error signal U1, and this is regarded as the position error signal U2, and the positioning of the head is controlled. At this time, in the conventional run-out correction, the position error signal U2 is generated by executing the calculation of the equation (14) for each predetermined discrete period while performing a table lookup of the cos value and the like.

However, in the magnetic disk drive of this embodiment, since the calculation results of the second and subsequent terms on the right side of the equation (14) are recorded in the memory 16, the calculation processing time is shorter than in the conventional case, and , The position error signal U2 can be generated by adding the position information and the correction value. It should be noted that these correction values can be rewritten by periodic re-measurement to easily deal with aging, environmental changes, and the like.

Next, in the above description, the correction value is periodically re-measured. However, an appropriate periodic re-measurement is performed, and a run-out detection unit provided for the purpose of preventing data read / write malfunction due to run-out. The operation of No. 17 will be described. In this embodiment, since the purpose is to improve the data format efficiency and the memory efficiency, servo signals important for positioning the head 9 are:
The feature is that it is a discrete information signal having a certain period interval (sampling period). For this reason, if the run-out amount suddenly changes for some reason while the head 9 is moving on the data section during the acquisition of the next servo information during head positioning or in a so-called on-track state after the positioning is completed, The occurrence of a track cannot be sensed until the next servo signal is obtained.

Therefore, the head 9 continues the data read / write operation without knowing that the off-track has occurred, and the read or written data lacks reliability. Therefore, the run-out detecting section 17 reads the signal in the data area while performing the data read / write operation by the head 9, and always detects the temporal change of the run-out. Further, the detected value is compared with a reference value in the memory 21 by the comparator 22. If the detected value exceeds a preset reference value, the read / write operation by the head 9 to the upper interface circuit 14 is performed. Output the stop command. This reference value is measured in a final inspection step or the like in which the above-described run-out correction value is written, and is calculated by averaging.
Record in 1.

Next, the operation of the run-out detecting section 17 will be specifically described. Specific cylinder n on disk 23 in FIG.
(Not shown), the head positioning operation is completed, and the head 9 is stopped, that is, the positioning control for the head 9 in the on-track state is performed as shown in FIG.
The servo signal read by the head 9 itself is supplied to a peak hold circuit 12 via a read / write amplifier 11 to hold the peak value, and the A / D converter 13 discretizes (digitizes) the peak value. After that, it is supplied to the runout processing unit 2 in the microcomputer 15 and supplied to the adder 3 to be added to the target value input.

The actual position error signal U1 extracted from the adder 3 is added to the run-out correction value U3 from the run-out processor 2 to form a position error signal U2.
, And an appropriate control amount is calculated by the compensation unit 5, and D
/ A converter 6, and VC through an amplifier 7.
The position of the head 9 is controlled by being supplied as a drive signal to M8 and driving appropriately. The head 9 performs a data read / write operation in accordance with an instruction from the upper interface circuit 14, and at this time, the run-out detection unit 17 starts operating. The run-out detection unit 17 amplifies the read signal of the head 9 performing the data read / write operation by the read amplifier 18 shown in FIG. 2, and then holds the peak value by the peak hold circuit 19, and further holds the peak value. The high-frequency component is removed by passing the value through a low-pass filter 20, and the value is compared with a reference signal stored in a memory 21 by a comparator 22.

As a result of this comparison, if the detected peak value is equal to or smaller than the reference signal, the change amount of the run-out amount can be ignored, and it is determined that it is not necessary to update the run-out correction value, and the head 9 reads / writes data. Continue operation. On the other hand, if the detected peak value is larger than the reference signal,
It is determined that the run-out correction value needs to be updated, or that an unexpected disturbance has been mixed in one servo sector, and a command to immediately stop the read / write operation by the head 9 is output to the upper interface circuit 14.

Thereafter, the upper interface circuit 14
Indicates a retry operation for the read / write operation of the head 9. If a similar stop command is output from the run-out detection unit 17 during the retry operation,
The upper interface circuit 14 again instructs the run-out processing unit 2 to confirm the run-out correction value. Upon receiving this instruction, the run-out processing unit 2 executes the automatic run-out correction operation again, and updates the stored run-out correction value.

By the above-described series of operations of this embodiment, the positioning accuracy can be improved, and at the same time, the data read / write malfunction due to the deterioration of the runout can be prevented. In this embodiment, the runout measurement,
Since the calculation of each correction data and a series of calculations related to the actual correction can all be realized by firmware internal processing,
No special hardware is required for the automatic run-out correction, and the cost can be reduced. Further, the present invention can be easily applied to a small magnetic disk device of 2.5 inches or less, a removable disk device, and a High-FDD (FDD: floppy disk device).

[0056]

As described above, according to the present invention,
The run-out correction data corresponding to the servo signal required for head positioning is recorded in the memory, and by using the run-out correction data, an appropriate head position error signal corresponding to the eccentricity of the disk can be generated instantaneously. Therefore, it is possible to easily correct the undulation of the servo signal caused by the runout caused by the eccentricity of the disk rotation axis of the magnetic disk device, and to improve the accuracy of the adjacent track pitch.

Further, according to the present invention, since the change in the runout amount is always detected, even when an unexpected disturbance or a sudden increase in the runout amount can be detected instantaneously, thereby reading / writing data. Prevention of malfunction during operation.

Further, according to the present invention, runout measurement,
Since the calculation of each correction data and the series of calculations relating to the actual correction can all be realized by firmware internal processing, automatic run-out correction can be performed without preparing special hardware, thus reducing costs and reducing various costs. The present invention is applicable to a magnetic disk drive.

[Brief description of the drawings]

FIG. 1 is a block diagram of an embodiment of the present invention.

FIG. 2 is a block diagram of an example of a runout detector in FIG. 1;

FIG. 3 is a flowchart for explaining the operation of a runout processing unit in FIG. 1;

FIG. 4 is an explanatory diagram of zone division on a magnetic disk.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Positioning control part 2 Runout processing part 3, 4 Adder 5 Compensation part 6 D / A converter 8 Voice coil motor (VCM) 9 Head 10 Actuator mechanism 12, 19 Peak hold circuit 13 A / D converter 14 Upper interface circuit 15 Microcomputer (microcomputer) 16, 21 Memory 17 Run-out detector 22 Comparator

────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-4-170775 (JP, A) JP-A-6-103592 (JP, A) JP-A-3-216874 (JP, A) JP-A-7-107 249276 (JP, A) JP-A-10-21662 (JP, A) JP-A-6-96542 (JP, A) JP-A-5-282807 (JP, A) JP-A-8-293174 (JP, A) JP-A-5-144180 (JP, A) JP-A-4-368681 (JP, A) JP-A-1-292680 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) G11B 21/10

Claims (4)

(57) [Claims] 1. A runout correction which receives a head position signal reproduced by a magnetic head from a magnetic disk, measures a runout component synchronized with the rotation of the magnetic disk by an operation using a Fourier transform equation, and generates runout correction data. Data generation means; storage means for storing the run-out correction data; and a run-out signal from the storage means to a first position error signal between a target value input from a host device and a head position signal reproduced from the magnetic disk. Adding means for adding correction data, an actuator mechanism for controlling the position of the magnetic head to a target position using a signal obtained from the adding means as a second position error signal, and reading / writing data from / on the magnetic disk Previous
The runout component is always detected and the detected runout component is detected.
When it is determined that the minute is greater than a preset reference value
Stops the read / write operation by the magnetic head.
And a run- out detection unit.
Retry operation after read / write operation was stopped due to
Sometimes, the detected runout component is
If the runout correction data is determined to be
Measurement of the runout component by the
The generation of the naut correction data is performed again, and the
A head positioning control method for a magnetic disk drive, wherein the stored run-out correction data is updated . 2. The run-out correction data generating means,
2. The head positioning control method for a magnetic disk drive according to claim 1, wherein the magnetic disk is divided into a plurality of zones, and the same run-out correction data is generated in each zone. 3. The run-out correction data generating means,
3. The head positioning control method for a magnetic disk drive according to claim 1, wherein a runout component synchronized with the rotation of the magnetic disk is periodically measured to generate runout correction data. 4. A readout amplifier for amplifying a read signal of the magnetic head, a peak hold circuit for holding a peak value of an output read signal of the read amplifier, a runout detection unit, A low-pass filter for removing a high-frequency component, a memory storing the preset reference value, and comparing the output signal of the low-pass filter with the output reference value of the memory; head but when it is larger than the output reference value of said memory, a magnetic disk apparatus according to claim 1, characterized in that more a comparator for outputting an instruction signal for stopping the read / write operation by the magnetic head Positioning control method.